Anos de vida perdidos por morte prematura por causas externas: análise da tendência e uma proposta metodológica para redistribuição dos códigos-lixo

Abstract

Violence is considered a social issue, or even a social and historical phenomenon, which occurs in various forms of human relations and that can manifest itself in acts of individuals and institutions, ie, carried out by individuals, groups, nations, in order to cause any physical or psychological harm on others. The term violence in this study refers to deaths from external causes, including intentional and unintentional deaths, ie, assaults, suicides and accidents in general. The objective is to analyze the evolution of the mortality burden in the state of Rio de Janeiro and propose the reallocation of deaths by undetermined intent through a new methodology. The data used are from the Mortality Information System. In this study, it was used the indicator YLL (Years of Life Lost) to assess the causes of violent behavior over time in the period 1996-2009, for the great region health state of Rio de Janeiro, using mixed effects models. The multinomial logistic regression was applied in death data by known underlying cause, using informations such as injuries and individual characteristics of victims, to predict what would be the underlying cause of death undetermined in the records with similar characteristics to the deaths with known causes. The results show that violence has increased in the state, with emphasis on great region North. In the capital and metropolitan areas, there was a stabilization of rates, except for deaths due to assaults, that decreased, but YLL rates remained high. The two methods of relocation, the Burden of Disease Method and this new proposal, all of them increase mortality rates by group of causes, but the group that suffered the greatest impact was the one of death by falls. The results, though limited, suggest a proposal of combination of the two methodologies. For deaths with underlying cause of Y10 (Poisoning by and exposure to nonopioid analgesics, antipyretics and antirheumatics, undetermined intent) to Y33 (Other specified events, undetermined intent), the methodology would be the Burden of Disease Method, and for the deaths of Y34 (Unspecified event, undetermined intent), it would be used the proposed method.